Liquid-liquid extraction of copper(II), zinc(II), cadmium(II), and lead(II) from aqueous solution and sewage effluent using phenoxy-amino ligands

Water safety and quality are crucial to human development, well-being, and the ecosystem (Chowdhury et al. 2016; Zamora-Ledezma et al. 2021). Unfortunately, the available water resources have negatively been influenced by a continuously increasing population, rapid industrialization, increasing urbanization, and careless utilization of natural resources (Azimi et al. 2017; Vardhan et al. 2019). Besides over-exploitation, direct or indirect discharge of various organic and inorganic pollutants from industries into water bodies without adequate treatment makes access to safe drinking water worse even in water-rich areas (Ali et al. 2016; Renu & Singh 2017). While microbiological contamination of water sources is an issue in many countries, emphasis on pathogens tends to steal the focus away from inorganic pollutants (Martínez-Santos 2017). Inorganic pollutants of great concern are heavy metals, which can be present in water as a result of both natural and anthropogenic processes (Renu & Singh 2017). Poorly treated industrial and agricultural wastewater contains high concentrations of heavy metals, which are often discharged into the natural water bodies (Chowdhury et al. 2016). Heavy metals are generally defined as those metals that possess a specific density of more than 5 g/cm³ and are toxic even at low concentrations (Carolin et al. 2017). Examples of heavy metals of most concern include lead (Pb), zinc (Zn), copper (Cu), arsenic (As), cadmium (Cd), chromium (Cr), nickel (Ni), and mercury (Hg) (Ahmed & Ahmaruzzaman 2016; Hu et al. 2021).

The presence of heavy metal pollutants in the water environment has been of great concern because of their established negative effects on the health of humans and ecosystems (Jaskuła et al. 2021). At trace levels, some of these heavy metals such as copper and zinc are essential elements that play important roles in human metabolism (Lamsayah et al. 2016; Magu et al. 2016), although they can be toxic at high concentrations. Other heavy metals, such as cadmium, mercury, and lead have no known essential role in living organisms and are toxic at even trace concentrations (Magu et al. 2016). The common heavy metals that have been identified in polluted wastewater include arsenic, copper, cadmium, lead, chromium, nickel, mercury, and zinc (Akpor & Muchie 2010). Exposure to these heavy metals occurs through absorption, inhalation, and ingestion. Ingestion through drinking water has been reported as one of the major heavy metal exposures (Mahdavi et al. 2018). Removing heavy metals from wastewater is necessary because heavy metals' exposure can exert several adverse health effects such as chronic and sub-chronic ones that include shortness of breath and neurotoxic, mutagenic, and teratogenic effects with various types of cancers, which depend on the heavy metal type (Chowdhury et al. 2016; Mahdavi et al. 2018).

Removing heavy metals from industrial effluents before discharging is an important issue for health and environmental safety and is an important step toward safe drinking water (Da'na 2017). Thus, due to the noxious effects of these heavy metals and their persistence in the environment due to their non-degradable nature (Ayangbenro & Babalola 2017), research is needed to ensure safe drinking water. Many conventional approaches such as ion exchange, membrane filtration, chemical precipitation, coagulation-flocculation, adsorption, and electrochemical methods have been developed for the remediation of these noxious toxicants from contaminated wastewater (Sayin et al. 2018; Ai et al. 2020a, 2020b, 2021). However, these technologies suffer from several limitations including incomplete removal of metals, use of expensive reagents, high energy consumption, sensitive working environments, and production of toxic sludge, and, therefore, there is an urgent need for more practical and efficient technologies (Uysal Akku¸ et al. 2015; Zamora-Ledezma et al. 2021).

Liquid-liquid extraction is emerging as an alternative remediation strategy because it provides the choice of extractant type that influences the selectivity and extraction capabilities toward heavy metals even at trace levels (Karakuş & Deligöz 2015; Lamsayah et al. 2016; Sarıöz et al. 2018; Sayin et al. 2018). Most of the existing chelating agents, although simple and easy to synthesize, have not been designed to selectively bind heavy metals of interest and do not form stable complexes (Matlock et al. 2002b). Consequently, the formed complexes decompose, releasing the heavy metals back into the environment. Examples of the widely used commercial reagents include sodium dimethyldithiocarbamate (SDTC) and sodium thiocarbonate (STC) (Figure 1) (Matlock et al. 2002a). Unfortunately, laboratory experiments have shown that these compounds can decompose into hazardous materials during chelation, have high leaching rates, and are ineffective at pH less than 4.0 (Matlock et al. 2001, 2002a). According to the literature data on SDTC and STC, the reagent–metal combinations can readily decompose into other substances, including HgS (Matlock et al. 2001).

The purification of water remains one of the major challenges in modern society because the provision of clear and safe domestic water is vital for mankind. Thus, the development of chelating ligands that can remove heavy metals from water is of great significance. Through careful modification of the ligand moiety, the electronic and steric properties can be fine-tuned to target specific metal ions. Ideally, these ligands should work quickly, have a high binding capacity for the target analytes, and should not release their bound toxic metal ions easily. The focus has fundamentally to be on the design of chelating ligands that are chemically specific and capable of permanently sequestering toxic heavy metal ions from wastewater discharges, even at trace levels. It is against this background that the performance of phenoxy-amino ligands (L1 and L2) in the extraction of selected heavy metals from aqueous solutions was investigated in this study. The effects of the nature of metal ions, ligand structure, and contact time have been investigated and discussed. The presented extraction protocol was extended to sewage effluent. The capacities of the phenoxy-amino ligands in the extraction of the metal ions were determined by atomic absorption measurement.

The solvents were obtained from Merck and dried or distilled using appropriate methods. The chemicals; salicylaldehyde (98%), N,N-(diethyl)ethylene-diamine (99%), and 2-aminoethanethiol were obtained from Sigma Aldrich and used without any further purification. The starting materials, 2-[(E)-{[(2-diethylamino)ethyl]imino}methyl]phenol (S1), and 2-((E)-(2-mercapto-ethylimino)methyl)-phenol (S2), and the ligands, 2-(((2-(diethylamino)ethyl)amino)methyl)-phenol (L1) and 2-(((2-mercaptoethyl)amino)methyl)phenol (L2) were prepared according to the methods cited in literature (Nyamato et al. 2015; Ngcobo & Ojwach 2017). All ¹H NMR and ¹³C{¹H} NMR (100 MHz) spectra were collected on a 400 MHz Bruker Ultra shield NMR spectrometer using deuterated solvent, i.e. CDCl₃. The infrared spectra were collected on Agilent Technologies, Cary 630 FTIR, and Perkin Elmer Spectrometer 100. Mass spectrometric data collection was performed using an LC Premier micro-mass Spectrometer model LCMS-2020. Determination of the compounds' elemental compositions was done on a Thermal Scientific Flash 2000, all at the University of KwaZulu-Natal. Metal concentrations were measured using atomic absorption spectrophotometry on a Shimadzu Atomic Absorption Spectrometry (AAS) 6300 at the Kenya Industrial Research and Development Institute (KIRDI), Nairobi.

The extraction of metal ions was carried out using equal volumes (20 mL) of 1,000 ppm metal nitrate solution (aqueous phase) and dichloromethane solution containing an equivalent amount of the appropriate ligand, which was transferred into a 100 ml conical flask. Solvent extractions were carried out by stirring the mixture for 2 h. The temperature was maintained constant at 25 °C during all experiments at pH 7.00. After the 2 h period, the aqueous phase was collected, appropriately diluted with distilled water and the extraction efficiency of each ligand was determined by the decrease of metal ion concentration in the aqueous phase by AAS. Moreover, other samples were collected and analyzed at time intervals of 0.5 h, 1 h, and 24 h. Each individual solvent-extraction experiment was carried out in triplicate and the arithmetic means were determined.

One-liter (1 L) water samples were collected at the effluent release point of the Embu sewage treatment plant. The treatment plant receives wastes that emanate from, among other sources, learning establishments, garages, and small microenterprises located in the vicinity of the treatment plant that drains into the treatment plant due to surface runoff. The wastewater is then treated up to the tertiary stage before being released to flow into river Rupingazi. The samples were divided into 250 mL aliquots and stored at 4 °C in tightly capped plastic bottles. The levels of zinc(II), cadmium(II), lead(II), and copper(II) ions in the water sample were determined by following the literature procedure (Sayo et al. 2020). Heavy metals extraction was done by adding 50 mL of the ligand solution (5.0 × 10⁻³ M) in dichloromethane to 250 mL of unfiltered water sample and the mixture was then shaken for 2 h. The aqueous phase was then separated, filtered, digested by following the literature procedure (Sayo et al. 2020), and then analyzed using AAS.

Treatment of the 2-[(E)-{[(2-diethylamino)ethyl]imino}methyl]phenol (S1) and 2-((E)-(2-mercaptoethylimino)methyl)phenol (S2) with excess amounts of NABH₄ produced the respective phenoxy(imino) ligands, 2-(((2-(diethylamino)ethyl)amino)methyl)phenol (L1) and 2-(((2-mercaptoethyl)amino)methyl)phenol (L2), respectively, in quantitative yields (Figure 2). The ligands were fully characterized by ¹H NMR, ¹³C{¹H} NMR and FT-IR spectroscopy, mass spectrometry, and elemental analyses. ¹H NMR spectroscopy was particularly instrumental in the structural elucidation of these amine ligands, and their spectra showed all the signature peaks expected of these reduced compounds. For example, the absence of the imine peak at between 8.31–8.45 ppm and the emergence of new singlets attributed to CH₂ protons at about 4.00 ppm in the ¹H NMR spectra of the reduced ligands, L1 and L2, confirmed the successful reduction of the imine ligands to their corresponding amine compounds, as shown in the supplementary information (Figures S1–S4).

Moreover, IR stretching frequencies of 3,303 and 3,298 cm⁻¹ for these phenoxy(amino) ligands L1 and L2, respectively, were indicative of the formation of secondary amines (Figures S5 and S6). Further analysis by mass spectrometry of L1 and L2 produced molecular ions associated with the formulae in Figure 2. Elemental analysis data were in agreement with the proposed empirical formulae (Figure 2) and confirmed the purity of these ligands.

The ability of 2-[(E)-{[(2-diethylamino)ethyl]amino}methyl]phenol (L1), and 2-((E)-(2-mercaptoethylamino)methyl)phenol (L2), to remove zinc(II), cadmium(II), lead(II), and copper(II) from aqueous media was investigated by liquid-liquid extraction using the dichloromethane-water biphasic system.

The identity of the metal ion had a significant effect on the extraction capabilities (Table 1). For instance, Cu(II) had the highest binding affinity for L1 at 97% followed by Zin (II) (91%), Cd(II) (76%), and Pb(II) (67%) under similar experimental conditions. The extraction efficiency of L2 was, however, greatest for cadmium(II) at 92% followed by copper(II), zinc(II), and lead(II) with binding affinities of 78, 75, and 59%, respectively.

The overall extraction efficiency for L1 was observed to be in the order of copper(II) > zinc(II) > cadmium(II) > lead(II) which is consistent with the Irvin-William series (Njoroge et al. 2013). Irvin-William series describes an empirical increase in stability of M²⁺ octahedral complexes as a function of atomic radius. Thus, based on Irvin-William series, Cu(II) forms stable complexes because of its relatively small ionic radius. On the other hand, the Hard-Soft Acid-Base (HASB) theory dominated the binding ability of L2. According to this theory, the degree to which a metal ion will bind to a ligand is greatly influenced by the metal ion's chemistry and preference for covalent or ionic bonding. The metal ions, in this case, behave as Lewis acids by accepting a lone pair of electrons from the ligands, which act as Lewis bases. Both copper(II) and zinc(II) are intermediate acids, while cadmium(II) is a soft acid according to the HSAB principle. It is thus conceivable that the introduction of a soft donor (S) atom in L2 results in moderation of the donor abilities of the phenoxy-amino ligand. It is, therefore, reasonable that the better binding efficiency of L2 to cadmium(II), compared to its affinity for the borderline hard acids copper(II), zinc(II), and lead(II) ions could be attributed to the ligand modification arising from the substitution of a hard N-donor atom with a soft S-donor atom.

The extraction efficiencies of L1 and L2 for zinc(II), cadmium(II), lead(II), and copper(II) are remarkably higher compared to those of their corresponding phenoxy-imino ligands in our previous report (Wambugu et al. 2021). In particular, the extraction efficiency of L2 for cadmium(II) (92%) is higher than that of its corresponding phenoxy-imino ligand of 87% and remarkably higher compared to those from our previous reports for (3,5-dimethylpyrazol-1-yl)ethanol (Njoroge et al. 2013) (15%) and (pyrazolylmethyl)pyridine (Ojwach et al. 2012a, 2012b) ligands (31%). However, the good performance of L1, which contains the hard N-donor atom, in the extraction of cadmium(II) implies that the Irving-William series plays a more predominant role compared to the HSAB theory.

The nature of ligand structure was also noted to influence the extraction efficiency of ligands L1 and L2 (Figure 3). For example, greater extraction efficiency was observed for L1 containing a borderline N-donor atom (Et₂N group) towards copper(II) (97%) and zinc(II) (91%). This could be attributed to the stronger coordination of the N-donor atom to the divalent metal ions, the majority of which are borderline hard acids (hard-soft acid-base theory).

A similar observation was made by Griesser et al., in a report of the study on the binding properties of copper(II) towards 2,2-bipyridyl ligand where the greater chelating effect of the ligand was attributed to the remarkable stability of the copper(II) and nitrogen bond (Griesser & Sigel 1970). Moreover, the substitution of the N-donor atom in L1 by a soft S-donor atom in L2 resulted in a decrease in the binding affinity for the borderline hard acids copper(II), zinc(II), and lead(II), from 97% to 78%, 91% to 75, and 67% to 59%, respectively.

Nevertheless, the substitution of a borderline N-donor atom (Et₂N group) in L1 by a soft S-donor significantly increased the extraction efficiency for cadmium(II) from 76% to 92% which could be attributed to the moderation of the donor abilities of L2 by the soft S-donor atom resulting in greater binding affinity for the cadmium (II) soft acid. This trend mirrors the earlier results reported for its corresponding phenoxy-imino ligand (Wambugu et al. 2021). Similar observations have been reported in a related study on remediation of cadmium and lead polluted soil using thiol-modified biochar by Fan et al., (Fan et al. 2020) where cadmium(II) was selectively adsorbed over lead(II). It is, therefore, reasonable to argue that the improved extraction performance of L2 for cadmium(II) compared to its affinity for the borderline hard copper(II), zinc(II), and lead(II) ions could be attributed to the ligand modification arising from the substitution of a borderline N-donor atom (Et₂N group) with a soft S-donor atom.

The influence of contact time on the extraction abilities of L1 was also investigated (Table 2). This was done by varying the reaction time from 0.5 to 24 h.

It is evident from the results that percentage extraction is affected by time variation. Increasing the mixing time from 30 min to 24 h resulted in a significant increase in the extraction efficiency of all the metal ions. For instance, there was a significant increase in the extraction of copper(II) from 50% at 0.5 h to 97% at 2 h. Further increase in mixing time from 2 to 24 h resulted in increased extraction efficiency of zinc(II), cadmium(II), and lead(II) from 91% to 99%, 76% to 99%, and 67% to 92%, respectively. Nevertheless, the extraction efficiency for copper(II) was invariable during this period. Similar trends have been reported previously for (3,5-dimethyl-1H-pyrazol-1-yl)ethanol in the extraction of cadmium(II) (Njoroge et al. 2013), and for 2-pyridyl-N-(20-methylthiophenyl)-methyleneimine (PMTPM) during removal of mercury(II), cadmium(II), and lead(II) from water (Pobi et al. 2019) where the extraction performance of the respective ligands was observed to increase with time.

After the preliminary experiments, the extraction was extended to sewage effluent samples collected from the Embu sewage treatment plant (ESTP) using L1. The concentrations of copper(II), cadmium(II), zinc(II), and lead(II) in the effluent were determined as 1.32 ppm, 0.45 ppm, 5.12 ppm, and 3.20 ppm, respectively, at a pH of 6.9 (Table 3). The extraction efficiencies of 95, 78, 90, and 60% for copper(II), cadmium(II), zinc(II), and lead(II), respectively, were obtained.

The trend of an extraction efficiency of copper(II) > zinc(II) > cadmium(II) > lead(II) is consistent with the trend observed for the aqueous metal salts (Table 1) where the Irvin-Williams series dominated in the results for L1. The final concentrations of copper(II) and zinc(II) in the sewage effluent of 0.10 ppm and 0.51 ppm, respectively, were well under the permissible limits for irrigation water of 0.2 ppm for copper(II) and 2 ppm for zinc(II) (Sayo et al. 2020), suggesting the potential application of the chelating agent in the removal of these heavy metals from sewage effluent.

This paper describes the extraction properties of phenoxy-amino ligands 2-(((2-(diethylamino)ethyl)amino)-methyl)phenol (L1) and 2-(((2-mercaptoethyl)amino)methyl)phenol (L2) toward copper(II), zinc(II), lead(II), and cadmium(II). We have demonstrated that L1 and L2 show good potential as chelating agents for the removal of the heavy metals copper(II), zinc(II), lead(II), and cadmium(II) from water with a high preference for copper(II) cation. The extraction protocol was successfully applied for the removal of heavy metals from real wastewater samples. The results demonstrated the potential application of the phenoxy-amino ligands in the extraction of copper(II) and zinc(II) from sewage effluent. The nature of the metal ion, ligand architecture, and the Irvin-Williams series influenced the extraction ability of the ligands.

All relevant data are included in the paper or its Supplementary Information.